Authors: Robert L. Kosut; Robert R. Bitmead

Abstract: An explicit constraint representing a standard whiteness test is added to the framework of least-squares ARX identification in order to provide a means for model order selection without relying on a second data set. This is accomplished by a convex approximation of the non-convex standard whiteness test, thereby forming a convex optimization problem.

Keywords: system identification for control design,model uncertainty estimation

Identifier: 10.3182/20060329-3-AU-2901.00166

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Brett Ninness; Adrian Wills

Abstract: This paper describes a Matlab (or Octave) based software package for the estimation of dynamic systems. It has been developed primarily as a vehicle for profiling novel approaches relative to existing methods within a common software framework that streamlines comparisons. Key features of the toolbox include simplicity of use (particularly via automated entry of unspecified values), the support of a wide range of scalar and multivariable model structures which include certain nonlinear classes such as bilinear and Hammerstein--Wiener, the ability to handle both time and frequency domain data, the hand optimisation of certain key routines compiled against ATLAS libraries for optimum speed, the use of non-standard optimisation methods based on adaptive subspace gradient search and the Expectation-Maximisation method, and the fact that the toolbox is freely available from http://sigpromu.org for non-commercial use.

Keywords: parameter estimation,system identification,software

Identifier: 10.3182/20060329-3-AU-2901.00146

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Honghui Xu; Soura Dasgupta

Abstract: In an earlier paper we had presented a novel dual channel identification approach for mobile wireless communication systems. Unlike traditional channel estimation methods that rely on training symbols, this approach used a bent-pipe feedback mechanism requiring the mobile station (MS) to send portions of its received signal back to the Base Station (BS) for wireless channel identification. Using a filter-bank decomposition concept, we introduced an effective algorithm for identifying both the forward and the reverse channels based only on this feedback information. This new method permits transfer of computational burden from the MS to the resource rich BS and leads to significant savings in bandwidth consuming training signals. This paper proposes a more informative feedback method leading to significant performance improvement over our earlier scheme.

Keywords: identification,channel,feedback,subspace

Identifier: 10.3182/20060329-3-AU-2901.00080

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Peter C. Young

Abstract: The paper describes an optimal Instrumental Variable (IV) algorithm for estimating an AutoRegressive Moving Average model of a time series. This IVARMA method is based on a modification of a previous algorithm and utilizes the Simplified Refined Instrumental Variable (SRIV) algorithm to estimate the ARMA model from the results of initial, high order, AutoRegressive (AR) model estimation. Using Monte Carlo simulation, the new algorithm is compared with the maximum likelihood method of ARMA estimation, using the well known PEM algorithm, and shown to produce parameter estimates with similar, statistically efficient properties. It is also incorporated in the Refined Instrumental Variable (RIV) algorithm to produce a new implementation of RIV for the full Box-Jenkins TF model form. Once again, MCS analysis confirms that this performs in a similar, statistically optimal manner to PEM, without the need for gradient-type optimization and with less sensitivity to the choice of initial conditions.

Keywords: ARMA estimation,high order autoregression,optimal instrumental variable estimation

Identifier: 10.3182/20060329-3-AU-2901.00061

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Lennart Ljung; Qinghua Zhang; Peter Lindskog; Anatoli Iouditski; Rajiv Singh

Abstract: The paper describes additions to the Matlab System Identification Toolbox, that handle also the estimation of nonlinear models. Both structured grey-box models and general, flexible black-box models are covered. The idea is that the look and feel of the syntax, and the graphical user interface should be as close as possible to the linear case.

Keywords: nonlinear system identification,neural networks,nonlinear models

Identifier: 10.3182/20060329-3-AU-2901.00148

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Andrea Lecchini; Michel Gevers; Jan Maciejowski

Abstract: Iterative Feedback Tuning (IFT) is a data-based method for the tuning of restricted-complexity controllers with a standard H2 criterion which in general gives no a priori robustness guarantees. In this paper we elaborate on Loop Transfer Recovery (LTR) LQG synthesis techniques designed to achieve robustness of the feedback loop. We propose an IFT procedure that achieves approximate LTR and its associated robustness. The proposed procedure is illustrated with a numerical simulation example.

Keywords: iterative feedback tuning,robust control,loop transfer recovery

Identifier: 10.3182/20060329-3-AU-2901.00176

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Peter Young; Hugues Garnier; Marion Gilson

Abstract: The paper describes and evaluates an optimal instrumental variable method for identifying hybrid continuous-time transfer function models of the Box-Jenkins form from discrete-time sampled data, where the relationship between the measured input and output is a continuous-time transfer function, while the noise is represented as a discrete-time AR or ARMA process. The performance of the proposed hybrid parameter estimation scheme is evaluated by Monte Carlo simulation analysis.

Keywords: continuous-time models,hybrid models,instrumental variable,sampled data,parameter estimation

Identifier: 10.3182/20060329-3-AU-2901.00030

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Gentiane Venture; Yoshihiko Nakamura; Katsu Yamane

Abstract: This paper presents an original use of the non-linear least-squares method applied to the muscle dynamics of the human body. The human body dynamics is very complex because of the number of degrees of freedom and of the number of muscles, moreover the behavior of muscles is non-linear and subject specific. A dynamic model of muscle, commonly used by the biomechanics community, which is presented, gives a relation between muscle force, activity, length and velocity. An application to the flexion/extension of the joint elbow using four muscles is then proposed. The dynamic parameters of those four muscles are estimated experimentally by the non-linear least square method. The activity (input of the dynamic model of the muscle) is measured using electromyography. The human arm dynamics is analyzed in a motion capture studio which acquisition of movements allows to compute the inverse kinematics and the inverse dynamics. Finally the muscle force is estimated (input of the dynamic model of the muscle).

Keywords: musculo-tendon dynamics,inverse dynamics,musculo-skeletal human model,electromyogram,motion capture

Identifier: 10.3182/20060329-3-AU-2901.00188

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Alessandro Chiuso

Abstract: Subspace identification for closed loop systems has been recently studied by several authors. Even though results are available which allows to compute the asymptotic variance of the estimated parameters for several algorithms, less clear is the situation as to relative performance is concerned. In this paper we partly answer this last question showing that the SSARX algorithm introduced by Jansson, which requires preliminary ARX modeling, and its "geometric version" called PBSID in the literature, which does not require any preliminary estimation step, are asymptotically equivalent. The question as to which is to be preferred in practice when working with finite data size remains open.

Keywords: closed-loop identification,subspace methods,asymptotic properties,variance

Identifier: 10.3182/20060329-3-AU-2901.00042

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: None

Abstract:

Keywords:

Identifier: 10.3182/20060329-3-AU-2901.00225

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Christian P. Robert

Abstract: The toolbox available in Bayesian Statistics has increased considerably in the past decade and it has opened new avenues for Bayesian inference, the first and foremost being Bayesian model choice. The MCMC and particle filter technologies have hugely increased the potential for Bayesian applications, in particular in missing variable models, as illustrated in this short tutorial. We will also mention a new direction in this field, namely the development of adaptive algorithms that avoid a lenghty tuning to fit the problem at hand by automatically modifying the parameters of the algorithm.

Keywords: adaptivity,Bayesian inference,MCMC algorithm,Monte Carlo techniques,missing variables,model choice,particle filter,population Monte Carlo

Identifier: 10.3182/20060329-3-AU-2901.00004

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Attila Magyar; Denes Petz; Katalin M. Hangos

Abstract: In this paper a Bayesian approach for estimating the state of a qubit is proposed. It consists of two phases. First a component-wise separate Bayesian estimate of the Bloch vector components is calculated in the form of β-distributions. Then a regularization step is performed to respect the constraint that the Bloch vector must be in the unit ball. The properties of the proposed algorithm are investigated by simulation.

Keywords: system identification,Bayesian estimation,quantum systems,regularization

Identifier: 10.3182/20060329-3-AU-2901.00151

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Frida Eng; Fredrik Gustafsson

Abstract: This work investigates how stochastic unmeasureable sampling jitter noise affects the result of system identification, and proposes a modification of known approaches to mitigate the effects of sampling jitter. By just assuming conventional additive measurement noise, the analysis shows that the identified model will get a bias in the transfer function amplitude that increases for higher frequencies. A frequency domain approach with a continuous time system model allows an analysis framework for sampling jitter noise. This leads to a bias compensated (weighted) least squares algorithm. A continuous time output error model is used for numerical illustration.

Keywords: system identification,stochastic systems,least-squares estimation,maximum likelihood,frequency domain

Identifier: 10.3182/20060329-3-AU-2901.00094

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Wee Lee Chia; Robert Bond Randall; Jerome Antoni

Abstract: A further development of the matrix mean differential cepstrum technique for blind identification of convolved mixtures in multiple-input-multiple-output mechanical systems is presented. The stability of the technique for application on the whitened response measurements can be improved with the use of the QR-Cholesky decomposition instead of the SVD-EVD set. Two simulated two-input two-output, two degree of freedom systems of differing damping coefficients are tested. The improved results with smoother frequency response functions of the system are presented and compared to prior results.

Keywords: blind identification,mean differential cepstrum,matrix differential equation

Identifier: 10.3182/20060329-3-AU-2901.00211

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Emeline Guiu; Manuel Rodrigues; Pierre Touboul; Jean-Jacques Loiseau

Abstract: MICROSCOPE is a scientific space mission which aims to verify the Equivalence Principle with an accuracy of 10-15, 100 times better than the best current lab experiences. Special attention has been paid to the accelerometric environment of the satellite in order to decrease all the disturbing accelerations. The payload is the core of a complex servo system that allows the drag compensation and the attitude control of the satellite, that generates reference signals for the instrument calibration and gives the Equivalence Principle data. This paper focuses particularly on the development of servo manoeuvres needed for the calibration.

Keywords: calibration,accelerometers,simulation,aerospace,estimation

Identifier: 10.3182/20060329-3-AU-2901.00152

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Lianming Sun; Akira Sano

Abstract: The identification problem is considered for OFDM communication channel suffering from multi-path interferences. It is shown that leakage errors arise due to the multi-path effects whose delay times exceed the guard interval. The property of leakage error is analyzed, and a new channel identification algorithm based on the leakage error estimation is proposed. It is clarified that the proposed algorithm can also work in a blind manner. Some numerical simulation examples illustrate its effectiveness.

Keywords: channel identification,OFDM,guard interval,leakage error

Identifier: 10.3182/20060329-3-AU-2901.00081

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Xavier Bombois; Marion Gilson

Abstract: This paper considers a recently introduced paradigm for optimal identification experiment design and extends the results to the case of an identification in a model structure which does not contain the true system.

Keywords: experiment design,prediction error identification for control

Identifier: 10.3182/20060329-3-AU-2901.00077

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Krister Jacobsson; Hakan Hjalmarsson

Abstract: Fluid flow models have turned out to be instrumental for analysis and synthesis of primal/dual congestion control algorithms which rely on aggregated information from a network path. In particular stability has been analyzed using such models. In network congestion control, validation experiments will with necessity be performed in closed loop since the communication protocol has to be active. Guidelines on how such experiments should be carried out in practice has until now been lacking in the literature. Departing from the theory of modeling for control, we refine a fluid flow model by augmenting the customary model of transport latencies, link price and source control with estimator dynamics and sampling properties. The impact of cross-traffic and changes in network configuration is incorporated as well. Furthermore, we analyze, from a closed-loop perspective, how the network should be excited when validating such models. The resulting identification framework is used for validating the derived model using packet-level experimental data from NS-2 simulations.

Keywords: communication protocols,communication networks,identification,closed loop,computer networks

Identifier: 10.3182/20060329-3-AU-2901.00139

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Miroslav Simandl; Ivo Puncochar

Abstract: A new unique general formulation and solution of the active fault detection problem for a discrete-time stochastic system are proposed. The optimal active detectors and the dual controller are designed by the appropriate criteria minimization using closed-loop information processing strategy. The considered formulation encompasses the following cases: active detection with an a priori given input signal generator, active detection, and finally dual control as active detection combined with control. The results are illustrated in the linear multiple models framework.

Keywords: fault detection,fault identification,detection systems,optimal control,optimal experiment design,dynamic programming

Identifier: 10.3182/20060329-3-AU-2901.00074

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Michel Fliess; Cedric Join; Hebertt Sira-Ramirez

Abstract: Recent advances in estimation theory permit a new approach to nonlinear black box identification, where a phenomenological model is replacing a precise mathematical description. Convincing simulations are provided for two examples: • the classic ball and beam system, • a large scale linear system, where our setting may regarded as a powerful alternative to model reduction.

Keywords: nonlinear systems,black box identification,model reduction,derivatives of a noisy signal,input-output representation,differential algebra

Identifier: 10.3182/20060329-3-AU-2901.00062

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Jochen Abfalg; Frank Allgower; Martin Fritz

Abstract: In this paper, the problem of simultaneously estimating the states and selected parameters of a diesel combustion engine air path is considered. Due to the high complexity of the nonlinear air path model, this problem is challenging, further complicated as decisive physical insight is available in the form of inequality constraints. We therefore provide a derivative-free estimation algorithm, combining constrained online optimization with an unscented Kalman filter covariance update formula. To keep generality, the algorithm is derived and examined for a general class of nonlinear constrained systems. Experimental results put to test, confirm the efficiency of the provided estimation algorithm.

Keywords: state estimation,parameter estimation,constraints,nonlinear systems,diesel combustion engines,air path,control,monitoring,fault diagnosis

Identifier: 10.3182/20060329-3-AU-2901.00224

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Linda Brus; Torbjorn Wigren

Abstract: A recursive identification algorithm for systems described by nonlinear ordinary differential equation (ODE) models is proposed. The ODE model is parameterized with coefficients of a polynomial in the state variables and inputs, that describes one component of the right hand side function of the ODE. This avoids over-parameterization problems. The model is then discretized with an Euler integration method. The algorithm exploits a Kalman filter, where the state variables needed in the right hand side function are derived by numerical differentiation. This approach makes a standard Kalman filter applicable to the identification problem. Contrary to a previously described RPEM algorithm, the proposed Kalman filter scheme cannot converge to false local minima of the criterion function. The proposed algorithm is therefore suitable for generation of initial values for the RPEM. The performance of the Kalman filter based algorithm is illustrated using a numerical example.

Keywords: nonlinear systems,identification,Kalman filter

Identifier: 10.3182/20060329-3-AU-2901.00159

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Per Sjovall; Tomas McKelvey; Thomas Abrahamsson

Abstract: Constrained identification of state-space models representing a structural dynamic systems is addressed. Based on physical insight, transfer function constraints are formulated in terms of the state-space parametrization. An example shows that a method tailored for this application, which utilizes the non-uniqueness of a state-space model, outperforms the classic sequential quadratic programming method in terms of robustness and convergence properties.

Keywords: system identification,state-space models,mechanical systems,MIMO,constraints,optimization problems

Identifier: 10.3182/20060329-3-AU-2901.00209

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Piet M. T. Broersen

Abstract: A continuous-time autoregressive spectral estimator is introduced that applies the principles of a discrete-time automatic equidistant missing data algorithm to unevenly spaced data. This time series estimator approximates the irregular data by a number of equidistantly resampled missing data sets, with a special nearest neighbor method. The ARMAsel-irreg algorithm estimates and automatically selects a discrete-time AR model from a number of candidates. This selected model often has a number of spurious high frequency poles, which are incompatible with the continuous character of the irregularly sampled signal. Those spurious poles can be eliminated, by transforming only the poles of the discrete time model with a positive real part to matching continuous-time poles. The estimated continuous-time spectra can be accurate at frequencies much higher than the mean data rate.

Keywords: autoregressive model,nearest neighbor resampling,slotting,spectral estimation,time series analysis,uneven sampling,order selection,parametric model

Identifier: 10.3182/20060329-3-AU-2901.00134

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Loic Cuvillon; Edouard Laroche; Hugues Garnier; Jacques Gangloff; Michel de Mathelin

Abstract: This paper presents a general methodology for identifying the dynamical part of the continuous-time model of an articulated arm including flexibilities dedicated to visual servoing. Based on this model, a H∞ control law is designed and implemented, allowing to reach high dynamics.

Keywords: system identification,continuous-time models,instrumental variable,visual servoing,robot arm,H˞control,flexible manipulator

Identifier: 10.3182/20060329-3-AU-2901.00204

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006